1. Digoxin Reduces 30-Day All-Cause Hospital Admission in Ambulatory Older Patients with Chronic Heart Failure and Reduced Ejection Fraction Ali Ahmed, Mihai Gheorghiade, Gregg Fonarow, Kanan Patel, Inmaculada Aban, Richard Allman, Jerome Fleg, Robert Bourge University of Alabama at Birmingham Veterans Affairs Medical Center Birmingham, AL ACC Late Breaking Clinical Trial Session March 11, 2013, San Francisco, CA

2. Presenter Disclosure Information DISCLOSURE INFORMATION: No relationships exist related to this presentation Ali Ahmed, MD, MPH Digoxin Reduces 30-Day All-Cause Hospital Admission in Ambulatory Older Patients with Chronic Heart Failure and Reduced Ejection Fraction Dr. Ahmed was supported in part by the National Institutes of Health through grants (R01-HL085561, R01-HL085561-S and R01-HL097047) from the National Heart, Lung, and Blood Institute (NHLBI)

3. The Digitalis Investigation Group (DIG) trial was supported by the NHLBI and the VA Cooperative Studies Program This article was prepared using a limited access dataset obtained from the NHLBI and does not necessarily reflect the opinions or views of the DIG Study or the NHLBI. Funding Disclosure Information

4. New Perspective on an Old Drug A Very Old Drug

5. Discovered Over 2 Centuries Ago An Account of the Foxglove and Some of its Medical Uses William Withering 1785

6. N Engl J Med 1993; 329:1-7 Patients whose digoxin was discontinued (in the placebo group) had a higher risk of worsening heart failure (HR, 5.9; 95% CI = 2.1 to 17.2; P<0.001) N=23 N=4 Improves Heart Failure Symptoms (The RADIANCE Trial)

7. N Engl J Med. 1997; 336: 525-33 Digoxin significantly reduced the risk of hospitalization due to heart failure by 28% during 37 months of average follow-up, but its effect on hospitalization due to all causes was more modest (a 8% reduction) Placebo (n=3403) Digoxin (n=3397) Absolute Risk Difference Hazard ratio (95% CI) P value Heart Failure 35%27%–8% 0.72 (0.66–0.79) <0.001 All-Cause 67%64%–3% 0.92 (0.87–0.98) 0.006 Reduces Risk of Hospital Admission (The DIG Trial)

8. N Engl J Med. 1997; 336: 525-33 HR = 0.99; 95% CI = 0.91–1.07; P = 0.80 Does Not Increase Mortality (The DIG Trial)

9. In 1997, FDA approved digoxin for use in heart failure Approved by the FDA

10. Recommended by Guidelines 2009 Focused Update Incorporated Into the ACC/AHA 2005 Guidelines for the Diagnosis and Management of Heart Failure in Adults JACC. 2009 doi:10.1016/j.jacc.2008.11.013 Digitalis can be beneficial in patients with current or prior symptoms of HF and reduced LVEF to decrease hospitalizations for HF Recommendation Class: IIa Level of Evidence: B

11. SOLVD (1991): 66% US Carvedilol (1996): 90% RALES (1999): 73% CHARM-Alternative (2003): 45% RAFT (2010): 35% EMPHASIS (2011): 27% over the subsequent decades…in part due to lack of effect on death and downgrade in guideline recommendations However, Use Declined

12. N Engl J Med 2009;360:1418-28 Yet, Heart Failure Remains the Leading Reason for Hospital Admission and Readmission Condition at Index Discharge 30-Day All-Cause Readmission Most Frequent Reason for Readmission All Medical21.0%Heart Failure (8.6%) Heart Failure26.9%Heart Failure (37.0%) All Surgical15.6%Heart Failure (6.0%)

13. March 2010 The New Health Care Reform Act Signed into Law Requires CMS to reduce payments to hospitals with excess readmissions, effective October 1, 2012…

14. October 2012 Medicare imposed about $300 million financial penalties against over 2,000 hospitals that had excessive readmission

15. Objective To examine the effect of digoxin on 30-day all-cause hospital admission in older, potentially Medicare- eligible, adults with heart failure and reduced ejection fraction in the main DIG trial

16. Digitalis Investigation Group (DIG) Ambulatory chronic heart failure (N=6800) – Ejection fraction ≤45% – Normal sinus rhythm – From United States and Canada – Randomized to receive either digoxin or placebo – During 1991-1993 – Followed for an average of 3 years – >90% on ACE inhibitors and >80% on diuretics 3405 (50% of 6800) were ≥65 years of age

17. Baseline Characteristics (1) Variables n (%) or mean (±SD) Placebo (n=1712) Digoxin (n=1693) P value Age (years)72 (5) 0.974 Female426 (25%)415 (25%)0.802 Non-whites194 (11%)180 (11%)0.514 Body mass index (kg/m 2 )26.2 (4.7)25.9 (4.5)0.040 Heart rate (per minute)78 (12) 0.445 Systolic BP (mm Hg)128 (20) 0.643 Serum creatinine (mg/dL)1.4 (0.4) 0.938 LVEF (%)29 (9) 0.855 Cardiothoracic ratio0.54 (0.08)0.54 (0.07)0.855 NYHA Class III-IV602 (35%)603 (36%)0.599

18. Baseline Characteristics (2) Variables n (%) or mean (±SD) Placebo (n=1712) Digoxin (n=1693) P value Heart failure duration (mos)30 (37)30 (38)0.625 Prior myocardial infarction1168 (68%)1154 (68%)0.969 Current angina pectoris489 (29%)465 (28%)0.476 Hypertension815 (48%)784 (46%)0.448 Diabetes mellitus517 (30%)488 (29%)0.379 Chronic kidney disease1038 (61%)1045 (62%)0.513 Dyspnea on exertion1323 (77%)1306 (77%)0.924 Dyspnea at rest386 (23%)358 (21%)0.323 4 or more symptoms/signs1411 (82%)1384 (82%)0.525 Pulmonary edema by x-ray266 (16%)286 (17%)0.283

19. Baseline Characteristics (3) Variables n (%) or mean (±SD) Placebo (n=1712) Digoxin (n=1693) P value Dose of study medication 0.125 mg/day433 (25%)426 (25%) 0.25 mg/day1197 (70%)1209 (72%)0.430 0.375 mg/day or higher71 (5%)48 (3%) Pre-trial digoxin use739 (43%)744 (44%)0.646 ACE inhibitors1605 (94%)1591 (94%)0.784 Diuretics1405 (82%)1374 (81%)0.493 Nitroglycerines788 (46%)768 (45%)0.697 Overall, ALL baseline characteristics of patients assigned to digoxin and placebo were balanced except for a slightly lower BMI among those assigned to digoxin

20. Kaplan-Meier Plot 30-Day All-Cause Hospital Admission Follow-up in Days

21. 30-Day Hospital Admission Due to All Causes Placebo (n=1712) Digoxin (n=1693) Absolute Risk Difference Hazard ratio (95% CI) P value 8.1%5.4%–2.7% 0.66 (0.51–0.86) 0.002 In the 30 days after randomization, in patients assigned to digoxin, the absolute risk and relative risk for all-cause hospital admission was reduced by an 2.7% and 34%, respectively

22. 60-Day and 90-Day All-Cause Hospital Admission Hazard ratio (95% CI)P value At 60 days0.76 (0.63–0.91)0.003 At 90 days0.75 (0.63–0.88)<0.001 The effect of digoxin on 30-day all-cause hospital admission persisted during 60 and 90 days after randomization, suggesting the early benefit of digoxin was not at the cost of later harm

23. Placebo (n=1712) Digoxin (n=1693) Absolute Risk Difference Hazard ratio (95% CI) P value 6.5%3.5%–3.0% 0.53 (0.38–0.72) <0.001 In the 30 days after randomization, digoxin reduced the risk of hospital admission due to cardiovascular causes by 47% 30-Day Hospital Admission Due to Cardiovascular Causes

24. Placebo (n=1712) Digoxin (n=1693) Absolute Risk Difference Hazard ratio (95% CI) P value 4.2%1.7%–2.5% 0.40 (0.26–0.62) <0.001 In the 30 days after randomization, digoxin reduced the risk of hospital admission due worsening heart failure by 60% 30-Day Hospital Admission Due to Worsening Heart Failure

25. 30-Day Mortality Hazard ratio (95% CI)P value All-cause0.55 (0.27-1.11)0.096 Cardiovascular0.64 (0.31-1.31)0.222 Progressive heart failure 0.22 (0.05-1.04)0.056 Although few deaths (n=34) occurred, they were numerically fewer in the digoxin group (0.7% vs. 1.3% for placebo)…

26. 30-Day Combined Outcomes Placebo (n=1712) Digoxin (n=1693) Absolute Risk Difference Hazard ratio (95% CI) P value 8.7%6.0%–2.7% 0.69 (0.53–0.88) 0.003 …consequently, the composite outcome of all-cause hospitalization or all-cause death at 30 days also was reduced substantially (by 31%) Only 4 patients were hospitalized because of suspected digoxin toxicity within 30 days of randomization, of whom 3 were from the digoxin group

27. Subgroup Analyses

28. Post hoc analysis of RCT Generalizability concerns – Ambulatory vs. post-discharge – Admission vs. re-admission – HFrEF vs. HFpEF – Not receiving beta-blockers – Not receiving aldosterone antagonists Study Limitations

29. Digoxin reduced the risk of 30-day all-cause hospital admission in ambulatory older adults with chronic systolic heart failure receiving ACE inhibitors and diuretics If these findings can be replicated in contemporary older heart failure patients discharged from hospital after acute decompensation, digoxin may provide an inexpensive tool to reduce 30-day all-cause hospital readmission Conclusions

30. “After all, in spite of opinion, prejudice or error, Time will fix the real value upon this discovery, and determine whether I have imposed upon myself and others, or contributed to the benefit of science and mankind.” As Sir Withering Predicted in 1785 Dr. William Withering (1741-1799)

31. Advance Access Online Publication March 11, 2013